Introduction to Power Density Fundamentals

A power supply’s size and efficiency are both closely related to its power density. The more an engineer understands about power density, how it is calculated, its limitations, and how those limitations can be overcome, the better prepared he or she will be to leverage the most advanced power-dense technology to meet the demanding demands of modern power management systems.

Why Power Density is Important

In the context of power management, power density is a measure of how much power can be processed in a given space or area. However, there is more to it than that.

Power density is defined as the amount of power processed per unit volume or area. The most common measurement units are watts per cubic meter (W/m3), watts per cubic inch (W/in3), watts per square meter (W/m2), or watts per square inch (W/in2).

It allows for more power to be processed in a smaller space while also improving system functionality at lower, rather than higher, system costs. Furthermore, power density and efficiency are inextricably linked in power-delivery applications, and efficiency is still a driving force in modern engineering. In fact, some consider higher power densities to be the pinnacle of power system design.

The details of correctly calculating power density can be quite detailed, but the fundamental concept behind estimating power density is quite simple. The vast majority of power density measurements are volume or area-based.

Volume power density is calculated by dividing the output power processed (Vo x Io) by the area formed by the board’s rectangular area (W) multiplied by the height of the tallest component on that board (x H). Power density in terms of area would simply be the power processed (Vo x Io) by the board’s rectangular area (L x W).

Trade-Offs  and Limitations in Power Dense Solutions

There are trade-offs and constraints in high power density solutions. It is critical to be aware of these issues in order to comprehend how progress toward higher power densities can be made.

There are very specific limiting factors for how much power density can realistically be improved, which include some of the following:

• Turn-on and turn-off losses
• Charge related losses
• Conduction losses
• Reverse-recovery losses
• Thermal issues

With the exception of thermal issues, these factors are related to converter power losses.

Keys to Achieving High Power Density

A general understanding of what limits power density helps a designer focus on how high power density can be achieved despite those constraints. Modern power management solutions with high power density share the following fundamental characteristics:

• A low-conduction and switching-loss switching element.
• A converter topology optimized for high switching frequencies in conjunction with an appropriate control method.
• More passive power and control method integration, often accomplished monolithically or through multichip module technology that integrates multiple semiconductor dies, reducing footprint.
• Excellent thermal performance, with as small a temperature gradient as possible between the silicon temperature and the outside cooling surfaces.

High Power Density Innovations

Power-system development gained unprecedented flexibility and creativity with the introduction of high-density power components. This gave designers previously unimaginable options for implementing power architectures. In addition, the use of high-density power components simplified the design of distributed-power solutions.

There are also attempts to reach higher power densities in multi kilowatt dc-dc converters through galvanic isolation, which demand isolated gate drive solutions. Isolated gate drives have several advantages over non-isolated solutions, including the ability to protect against voltage spikes and other transients and provide better EMI performance.

The choice between a centralized power-system design and a distributed-power design was frequently driven by cost considerations. Increasing demands for higher reliability, higher performance, and higher system availability, on the other hand, have recently introduced another cost factor into the equation. That is the price of reduced performance or downtime, respectively.

Furthermore, as the cost per watt of high-density component power falls, distributed-power architecture (DPA) is becoming more appealing for a broader range of applications. Component power is now more easily and affordably applied to the construction of distributed-power solutions.

DPAs have several inherent advantages over centralized systems. Accurate voltage regulation, which is required in some applications, is often difficult and costly to achieve in the latter. Inductance can also be a problem when loads are at different distances from the source and/or are dynamic. Regulation occurs at or near the load in the case of distributed power. The lengths of the output leads are reduced, which reduces inductance and improves transient response.

Furthermore, a DPA system typically distributes higher voltages with significantly lower current levels than a centralized system. This architecture minimizes power losses during distribution, resulting in smaller and less expensive power conductors. Thermal issues are also better managed. Heat is dispersed and hot spots are eliminated by dispersing power conversion throughout the system. This frequently eliminates the need for heatsinks or fans.

High-density power components make partitioning easier while also providing other benefits. Using fixed resistors, potentiometers, or digital-to-analog converters, most second-generation converters, for example, can be trimmed or programmed from 10% to 110 percent of their output-voltage set point (DACs). With such a wide output voltage range, it’s possible that all of the converters in a distributed system not only have the same footprint and pinout, but also have the same nominal output voltage.

Certain tasks, such as thermal management, hot-swapping, and mechanical design, become easier to perform as the number of load partitions increases. Furthermore, partitioning the load to the card level simplifies power-system assembly and simplifies the dc supply-bus design.

The Increasing Importance of Power Density

From on-board electric vehicle battery chargers to power supplies for IoT (Internet of Things) technology in the home, power density is critical. And, because power density is linked to efficiency and size, its significance is growing.